Next Article in Journal
Influence of Nano-Sized SiC on the Laser Powder Bed Fusion of Molybdenum
Previous Article in Journal
Development of Solution-Processed Perovskite Semiconductors Lasers
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

Synthesis and Magnetic Properties of Carbon Doped and Reduced SrTiO3 Nanoparticles

Marina V. Makarova
Andrey Prokhorov
Alexander Stupakov
Jaromir Kopeček
Jan Drahokoupil
Vladimir Trepakov
3 and
Alexander Dejneka
Institute of Physics AS CR, Na Slovance 2, 182 21 Prague 8, Czech Republic
Graduate School of Engineering Science, Akita University, 1-1 Tegatagakuenmachi, Akita 010-0852, Japan
Ioffe Physical-Technical Institute RAS, 194 021 St-Petersburg, Russia
Author to whom correspondence should be addressed.
Crystals 2022, 12(9), 1275;
Submission received: 15 August 2022 / Revised: 1 September 2022 / Accepted: 4 September 2022 / Published: 8 September 2022


We report on the studies of the synthesis, structural, and magnetic properties of undoped SrTiO3 (STO), carbon-doped STO:C, and reduced STO STO:R nanoparticles. Fine (~20–30 nm) and coarse (~100 nm) nanoparticles with a single phase of cubic perovskite-type structure were sintered by thermal decomposition of SrTiO(C2O4)2. Magnetization loops of fine STO:C and STO:R nanoparticles at low temperatures and an almost linear decrease in magnetization with temperature indicate the realization of a soft, ferromagnetic state in them, with a pronounced disorder effect characteristic of doped dilute magnetic semiconductors. Oxidation and particle size increase suppress the magnetic manifestations, demonstrating the importance of surface-related defects and oxygen deficiency in the emergence of magnetism. It was found that oxygen vacancies and doping with carbon make similar contributions to the magnetization, while complementary electron paramagnetic resonance, together with magnetization measurement studies, show that the most probable state of oxygen vacancies, which determine the appearance of magnetic properties, are charged F+ oxygen vacancies and C-impurity centers, which tend to segregate on the surface of nanoparticles.

1. Introduction

The emergence of magnetic properties in non-magnetic materials with non-magnetic defects and impurities attracts particular academic and technological interest. Unfolded theoretical and experimental studies of bulk and nanoscale non-magnetic oxides (e.g., [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22]) have found that the effective non-magnetic defects leading to magnetism are oxygen vacancies (VO) [1,3,9,13,14,15,16,17,18,19,20,21], related complexes, nonstoichiometry [3,5,12,23], and non-magnetic 2p- dopants (e.q. C, B, N or P) substituting for oxygen atoms (C@O) [4,5,6,10,11]. Moreover, it turns out that the appearance of magnetism is especially manifested and is a quite general property of nanosized objects with developed surfaces [13,14,24,25,26,27,28,29,30,31]. Particular interests of such researches are of highly polarized non-magnetic d0ABO3 perovskite-type ferroelectrics and related oxides (e.g., [1,3,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,26,27,28,29,32,33]), opening up new possibilities for simultaneous control of the degree of freedom of charge and spin in these widely used functional materials of oxide electronics.
The main micromechanism of the occurrence of ferromagnetism caused by oxygen vacancies (Vo) is most often associated with the appearance of relatively deep spin levels in the band gap caused by vacancies. Such levels can mostly capture one electron, while the second electron occupies the conduction band. In ABO3 perovskites, the presence of Vo promotes the appearance of Ta4+ and/or Ta3+ (Ti3+ and/or Ti2+) B- ions with a nonzero net spin [13,16,17,19]. An increase in the concentration of Vo and B- ions with nonzero spin on the surface and boundaries of grains and nanoparticles causes the formation of clusters, causing the possibility of a strong ferromagnetic exchange interaction.
The occurrence of magnetism caused by sp-elements doping, theoretically considered in [4,5,6,10,11,22], is much less investigated, especially experimentally. According to theory, magnetism induced by oxygen substitution with p-impurities occurs in the event that the orbital energies of their p-states are localized in the band gap of the host crystal, causing the possibility of spontaneous spin polarization, so the system gains magnetic semimetal or semiconducting properties [5,6]. The hole-mediated mechanisms of ferromagnetic interaction considered in [4] for ZnO with C@O carbon impurities should also be mentioned. Such substitution introduces O2p holes, which couple with the parent C2p localized spins by a pp interaction, similar to pd hybridization in semiconductors and oxides with TM impurities.
As far as we know, magnetism induced by p-impurities have been reported in a very limited number of experiments, in which carbon was used as an introduced impurity in C-implanted ZnO films and nanowires [4,30], TiO2 ceramics [8], and in TiO2/graphite carbon nanocomposites [31]. To the best of our knowledge, the possibility of the appearance of p-impurities induced magnetism in d0 ABO3 perovskite-like oxides has not been yet investigated experimentally.
In the present work, we report on the synthesis and study of the C-doped nanoparticles of STO, a comprehensively studied model archetype of the ABO3 perovskite-type ferroelectric oxides, with a wide range of technological applications. It is worth noting that synthesis of such material, the observation and estimation of the magnetic contribution introduced by carbon doping in STO, are a challenge. Synthesis of C-, N- and S- doped SrTiO3 has been reported for catalytic applications [34,35]. According to the DFT calculation [10], C impurity preferentially replaces Ti rather than Sr and O ions in STO, while spin-polarized states are formed only for C@O substitution; attainment of the concentration of C@O centers sufficient for the manifestation of magnetism is complicated. Moreover, such doping requires synthesis/treatment in reducing conditions (H2, CO atmospheres, solid carbon additions etc.) to remove lattice oxygen and substitute it with carbon. Such a process is accompanied by a partial removal of lattice oxygen and the formation of oxygen vacancies [35,36,37,38]. Since vacancies of VO themselves lead to the magnetic moment emergence, the problem of identification, estimation, and comparison of VO not substituted by carbon and C@O impurity contributions to magnetism becomes a complex task. It is also necessary to eliminate the possible presence of accidental contaminating impurities of magnetic cations in the materials under study. We planned to try to develop technology to synthesize and study the properties of the STO nanopowders doped with carbon (STO:C)-exhibiting magnetic properties. Particular attention was paid to the identification, evaluation and comparison of the magnetic contributions coming from VO and C impurities. For this purpose, nanoparticles of various sizes, subjected to additional thermal treatment in a reducing and oxidizing atmosphere, were synthesized and studied too.

2. Materials and Methods

The method of complex compound thermolysis was used for nanoparticle fabrication. Strontium titanyloxalate was chosen as the basic precursor, with a Sr:Ti:C ratio of 1:1:4 [39] so the carbon dopant gets uniformly introduced simultaneously with the phase formation. An excess of carbon was taken to ensure maximal possible introduction of carbon into the lattice, which may depend on particle size and annealing temperature. Such a scheme was expected to decrease both reaction temperature and particle size. Figure 1 demonstrates the used synthesis scheme. Ti(OCH(CH3)2)4 (Fluka, Buchs, Switzerland) was mixed with H2C2O4·2H2O (Lachema, Brno, Czech Republic) water solution in a 1:2 molar ratio. Concentrated NH4OH solution (Lachema, Brno, Czech Republic) was added until there was neutral reaction of the indicator paper to finalize titanyloxalate TiO(C2O4)22− formation. Then, Sr(NO3)2 (Sigma-Aldrich, St. Louis, MO, USA) solution was added to the titanyloxalate solution while stirring and then stirred for 30 min. The Sr(NO3)2 precursor was prior heated at 150 °C for 1 h to decompose possible hydrates. The white crystalline precipitate SrTiO(C2O4)2 was collected, filtered, washed with distilled water, and dried. Then, two parts were annealed in H2/Ar flow (both Linde, Dublin, Ireland, 20 sccm each) for 1 h at 700 or 1000 °C, yielding carbon-doped STO:C (in fact, reduced carbon-doped) «fine» and «coarse» powders, marked as Cf and Cc, respectively; we expected that the annealing at a higher temperature would result in a larger particle size. Another part was annealed at 700 °C in air for 1 h, to burn all the carbon dopant and obtain fine, undoped STO (marked as If) and then treated in the same reducing way as the initial precursor (see Figure 1, samples marked Rf and Rc for fine and coarse powders for low- and high- temperature annealing). Finally, fine C-doped powder was annealed in air for 1 h at 700 °C for oxidation (marked OCf).
The resulting powder phase composition and structure were inspected by X-ray diffraction (XRD) using the PANalytical X’pert diffractometer (Malvern PANalytical, Malvern, United Kingdom) at CoKα radiation, with a small addition of CoKβ. The obtained spectra were recalculated in CuKα and compared against the standard STO spectra and the spectra of possible phase admixtures. The lattice constants were derived from the spectra fit in a PowderCell software 2.4. by W. Kraus and G. Nolze (Berlin, Germany) using the Ag addition to the studied powder as an internal standard. A scanning electron microscopy (SEM) experiment was performed with a TESCAN FERA 3 scanning electron microscope (TESCAN, Brno, Czech Republic), equipped with an OXFORD Instruments X-max EDS SDD 20 mm2 detector (SEM-EDS) (Oxford Instruments, Abingdon, United Kingdom). To determine particle size distributions, 30+ individual nanoparticles were selected from SEM images and statistically analyzed. The same sets were used to calculate specific surfaces. Magnetic properties in the 10 K ÷ 300 K temperature region were characterized by the SQUID Magnetic Property Measurement System (Quantum Design) MPMS-XL 7T (San Diego, CA, USA).
The absence of accidental contamination of experimental samples with magnetic ions, such as Fe, Mn, etc., which could be a source of magnetism, was controlled by the electron paramagnetic resonance (EPR) method. Along this way, we also hoped to get data on the centers involved in the formation of the observed occurrence of actual ferromagnetism. Continuous wave EPR spectra were measured using a Bruker X/Q band E580 FT/CW ELEXSYS spectrometer (Bruker, Billerica, MA, USA) in the temperature range of 5–300 K. An ER 4122 SHQE Super X High-Q cavity with TE011mode was used. The samples were placed into quartz tubes with a diameter of 4 mm. The experimental parameters were: microwave frequency 9.8756 ± 0.0005 GHz, microwave power 1.500 mW, modulation frequency 100 kHz, modulation amplitude 0.2 mT, and the conversion time 60 ms. Neither magnetic admixture lines nor ferromagnetism were found in the starting materials.

3. Results and Discussion

Figure 2 shows the XRD spectra of the reference fine STO (If), reduced undoped (Rf and Rc), and reduced C-doped (Cf and Cc) nanopowders and standard peaks of STO marked with black bars. The designations and preparation details are given in the Figure 1 scheme and Table 1. A comparison of the standard XRD peaks of STO indicates that all the powders have a cubic perovskite-type structure, with some traces of SrCO3 admixture in If and Rf powders, which might appear due to slight Sr excess in the initial precursor.
Figure 3a presents an SEM image of a typical Cf powder. As it can be seen, the particles have a more or less spherical shape and are aggregated into flower-like structures, possibly inherited from the basic SrTiO(C2O4)2 crystalline precursor. Other fine powders have a morphology similar to that of Cf, as demonstrated in a Figure 3b–d. Particle size distributions are shown in the Figure 3e.
Particle size was roughly the same for all the powders obtained at 700 °C and slightly increased upon oxidization in Cf samples (OCf). Powders prepared at 1000 °C were coarser and had a broader particle size distribution due to more intensive crystallite growth. Thus, powders denoted as “f” or “fine” have ~20–30 nm particle size while “coarse” are ~100 nm. Table 2 shows the lattice constants and their relationship with particle size. It is seen that they do not largely differ between each other. However, slight lattice expansion was observed for the carbon-doped Cf particles, which is consistent with the data for Ti oxycarbide [40,41,42] and can be related to partial substitution of O ions with C (formation of C@O centers). In the oxidized C-doped powder OCf, the lattice constant did not undergo visible changes, probably because the oxidation temperature was insufficient for full decarbonization and affected only near-surface layers. Coarse powders, annealed at 1000 °C, have a lower lattice constant, compared to 25–30 nm powders, due to the effect of unit cell expansion in the fine powders [43,44]. The carbon-doping effect was not seen on them, probably because oxycarbide formation depends on particle size as well [36].
Figure 4 presents the results of magnetic studies for C-doped (Cf, Cc) and undoped (Rf and Rc) powders with fine (~20–30 nm) and coarse (~100 nm) particle size: (a) M–H hysteresis loops at T = 10 K and for C-doped at 300 K (Inset), (b) temperature dependence of magnetization for reduced and C-doped powders, and (c) saturation magnetization on nanopowder specific surface area at 10 K. It is seen that both coarse Rc and Cc nanopowders hysteresis loops are not unfolded, characterizing a slight nonlinearity of the M–H field dependence. At the same time, the M–H behavior of fine Cf and reduced Rf nanopowders clearly show unsaturated narrow hysteresis loops, which are very close to those observed in BaTiO3, SrTiO3, and KTaO3 crystals implanted with Co and Mn [45], exhibiting the properties of a soft ferromagnet. In Figure 4b, it is also seen that the magnetization of fine Rf and Cf is greater than that of the others, and, as in [45], almost linearly decreased with temperature, in contrast to the usual Brillouin temperature dependence that is most likely attributed to the disorder inherent in dilute magnetic semiconductors [46]. Similar behavior is observed for pure If and oxidized OCf fine powders, in which the observed magnetization is much less pronounced than for reduced Rf and Cf powders; hysteresis loops are very narrow and not unfolded and more meet the nonlinear M–H dependence. Figure 4a–c shows that the highest magnetization value is achieved in Cf and Rf nanoparticles with the minimum particle size and a large specific surface area, which indicates the importance of near-surface defects in the magnetism emergence [19,20,21]. Reasonably assuming that a certain fraction of oxygen vacancies in Cf particles is not replaced by carbon, we conclude that doping with both C and Vo induces close magnetization values and is efficient in the emergence of ferromagnetism.
Figure 4c demonstrates the increasing dependence of magnetization on specific surface areas in both reduced and oxidized powders, which may be related to the contribution of natural and artificial surface defects, oxygen vacancies, and dangling bonds. Ferromagnetism is almost suppressed in coarse Rc and Cc powders with low surface area, despite their thermal reduction and C doping. However, magnetization is pronounced in the fine nanopowders with large surface Cf, Rf, and If, where reduction and C doping additionally increases it in Rf and Cf, compared to reference If.
The M–H curve of C-doped nanopowder after thermal oxidation (OCf) significantly differs from the initial Cf powder at both 10 and 300 K (Figure 4a,b). Ferromagnetism almost disappears, while the particle size (Figure 3e) and surface area (Figure 4c) do not change much. After oxidation, the Cf powder changes color from black to white, which may indicate both carbon dopant elimination and VO filling in the resulting OCf sample. Taking into account that the EPR data presented below did not reveal any noticeable manifestations of the presence of Fe, Mn ions, and other magnetic impurities, which often contaminate perovskite-like oxides, we conclude that ferromagnetism in Cf and Rf is mainly due to “non-magnetic doping,” with carbon and with oxygen vacancies localized mainly in the near-surface regions of the nanoparticles. The resulting magnetic state is most likely the state of a soft ferromagnet, with the pronounced effect of the disorder characteristic of doped dilute magnetic semiconductors.
It should be noted that, concluding about the realization of a soft ferromagnetic state, it should be taken into account that ferrimagnets can also exhibit magnetic properties similar to those we observed. However, unlike ferrimagnets, STiO3 has a very simple structure in which the presence of various magnetic structural complexes in various crystalline environments leading to ferromagnetism seems extremely unlikely and has never been observed in SrTiO3 doped with magnetic impurities; it often exhibits the properties of dilute magnetic semiconductors (e.g., [45]). We do not know of any publications reporting the formation of a ferrimagnetic state induced by nonmagnetic defects and impurities, not only in strontium titanate and related ABO3 oxides, but also in oxides of simple crystal structure. Finally, from Figure 4a, it is seen that the hysteresis loops, the values of magnetization Cf and, especially important, Rf nanopowders with one type of defects (Vo) leading to magnetism are very close, which complicates the ferrimagnetic interpretation and allows us to assume a soft ferromagnetic state.
Figure 5 shows the broadband (Figure 5a) and short-range (Figure 5b) EPR spectra of the nanopowders under a study recorded at room temperature, and the short-range spectra for T = 25 K (Figure 5c,d). For all the Figures, the mass of the samples was fixed as 20 mg, with the exception of Figure 5a. It is seen that the EPR spectra at room temperature contain two wide bands that are at the noise level and will not be discussed, along with prominent pronounced absorption lines in the region of ~335 mT (g ≈ 2.002 A-lines). Figure 5b shows the detailed structure of A-lines, as well as the presence of another weak resonance in the region of g ≈ 1.978. Figure 5c presents the EPR spectra at T = 25, of which the most intense are the A-lines (Lorentzian shape with a linewidth of ~4 mT) of Cf nanopowder, and the weaker but quite pronounced are the A-lines of Cc and Rf.
The position of the EPR absorption A-lines in the region of g ≈ 2.002 meets the condition of resonance absorption with the spin s = ½, which can refer to a number of centers, such as itinerant electrons, which provide magnetic ordering [7,47,48], electrons in the conduction band and electrons of spin levels inside the energy gap, originating from the valence band upon doping with Vo, F-centers on the surface of nanoparticles, p impurities, and C dopants [4,5,14,31,49,50,51,52].
It is reasonable to assume that the A-lines of the EPR resonance of carbon-free fine reduced Rf nanopowders with the largest surface area are associated with oxygen vacancies, which, according to [9,14,15,20,53,54,55,56], are predominantly charged F+ centers, with a fairly deep localization, low formation energy, and a greater tendency to segregate on the STO surface than neutral triplet vacancy F-centers (with two electrons). The significant role of oxygen vacancies, especially surface ones, is convincingly confirmed by a strong decrease in the EPR signal and magnetization of carbon-free coarse Rc and a complete suppression of the A-EPR line of oxidized Cf particles, where thermal oxidation eliminates both oxygen vacancies and carbon dopants.
The most interesting feature of the EPR spectra in the region g ≈ 2.002 is the absorption of Cf nanoparticles, which turns out to be much stronger than that of other nanoparticles and is slightly shifted towards larger values of the g-factor, compared to the spectrum of Rf nanoparticles associated with oxygen vacancies, which agrees well with the results of the EPR studies of C-doped titanium dioxide [52]. This makes it possible to quite confidently relate the EPR spectrum of Cf nanoparticles in the region of g ≈ 2.002 with the carbon impurity centers. Another source of free electrons and the appearance of A–lines in the EPR spectra could be the phases formed by free residual carbon. However, the presence of similar ones in our case should be negligibly small, since they are not visible in the XRD spectra, even in the form of halos. In addition, the soft ferromagnetism inherent in the main allotropic modifications of carbon (diamond, graphite, nanographite, nanotubes, fullerenes) [57], if present, is negligible, much weaker than that observed in our experiments.
Thus, the results obtained provide sufficient grounds to believe that the observed narrow EPR lines in the g ≈ 2.002 region mainly originated due to oxygen vacancy- and carbon-containing defect complexes.
Table 3 arranges the room temperature magnetization values from Figure 4a,b and the intensities of the EPR A-lines (Figure 5b) in decreasing order.
It follows from Figure 4a and Figure 5c and Table 3 that the most pronounced ferromagnetism at room temperature and the intensity of the EPR A-line were observed in carbon-doped Cf and reduced Rf “fine” STO powders with the most developed surface. Note that the smallest value of magnetization was obtained in Cc nanopowders, while Cf nanopowders with the largest surface area revealed the highest magnetization value and the intensity of the EPR A-line spectra. That allows it to assume the leading role in the production of the magnetism of C-related defect as it enters segregating in the surface regions.
Correlation was also observed in carbon-free Rf and Rc powders, where the A-line could be certainly associated with VO. Both A-line intensity and magnetization were much higher for fine Rf, which can again testify that VO are preferably formed near the particle surface and are responsible for magnetic order emergence. This line is also absent in the reoxidized sample OCf after C and oxygen vacancies elimination, when magnetization disappears as well. These observations confirm the importance of the contribution of oxygen vacancies VO, most likely associated with charged F+ centers mentioned above, which have a low formation energy and a great tendency to segregate in the STO surface in magnetic order emergence.
Figure 5d shows the detailed resolved EPR spectra of a series of five weak resonance lines in the region of 334–354 mT (g ≈ 1.97) at 25 K, whose weak manifestation is noticeable already at room temperature (Figure 5b). With the highest probability, the found five resonance lines with g1 = 1.994, g2 = 1.978, g3 = 1.970, g4 = 1.953–1.957, and g5 = 1.922 correspond to the responses of different types of Ti centers. Ti3+ ions can be located as octahedral site ions in the distorted oxygen surrounding [58], in octahedral sites with two oxygen vacancies, and at Sr sites [59], and then the value g1 can be associated with Ti3+ adjacent to oxygen vacancies, with gx = 1.992 and gz = 1.994 obtained in the EPR spectra of reduced STO single crystals [59]. At the same time g2 and g3 lines can possibly be associated with Cr3+ ions, replacing Ti4+ (1.977, 1.974 [60,61,62], 9.780 [63]), an unavoidable impurity contaminating STO [64]. Note that the spectra of coarse Cc and Rc powders (Figure 5d) show individual EPR lines of Ti3+ EPR lines, while on fine Rf, these lines are blurred. This, again, may be explained by the larger role of the surface in Rf powder and by the reduction of coordination symmetry, thereby demonstrating again the connection between the occurrence of magnetic order and surface oxygen vacancies.
With regards to the properties of defects in the triplet (S = 1) state, they are studied scarcely, and their atomic nature is still tentative (see Refs. [65,66]). In this connection, the magnetic properties can be provided by triplet (S = 1) neutral oxygen vacancies (F0), where two compensating electrons of similar spin are localized at neighboring equivalent t2g states of neighboring Ti3+ ions. At the same time, the model was theoretically considered for STO [53,54,55,56] and experimentally for TiO2 [66], and could manifest itself in the EPR spectra of STO. However, the interpretation of EPR lines, based on the comparison with the g-factor values of rutile [65], has its limitations and requires caution. According to calculations, ionization energy of the triplet neutral F0 center in STO is very low (0.5 eV in the bulk and 0.2 eV on the surface). Therefore, the electrons of oxygen vacancy are weakly bound at Ti and most likely singlet. Due to low energy, F0 centers can be ionized into charged vacancies, such as F+ (with one electron V+O), having nonzero spin. This leads to the fact that the triplet oxygen vacancies turn out to be unstable, in contrast to the dominant F+. Note that the relative weakness of the EPR spectra associated with Ti ions clearly indicates the leading role of Vo and C-related defects in the production of the observed magnetism.

4. Conclusions

We developed a novel technology of carbon introduction into nanosized STO and studied the structure, magnetization, and EPR spectra of carbon-doped STO:C; we also reduced STO (STO:R) nanoparticles exhibiting the properties of a soft ferromagnet state with a pronounced disorder effect characteristic of doped dilute magnetic semiconductors, with ferromagnetism observed up to room temperature.
Particular attention was paid to the identification, evaluation, and comparison of the magnetic contributions coming from VO and C impurities. Despite the difficulty in estimating the concentration of defects in nanopowders doped with C, we concluded that the contribution to magnetism due to doping with C and VO was of the same order. It is important to note that the largest contribution to magnetism was made by C- and Vo- related defects segregating near the surface, which was accompanied by an increase in the magnetization of nanoparticles, with a decrease in their size and an increase in their surface area. Complementary electronic paramagnetic resonance and magnetization measurements made it possible to determine the EPR spectra connected to C impurities and charged oxygen vacancies F+ located in the region of g ≈ 2.002, which determine magnetic properties.

Author Contributions

Conceptualization, V.T.; methodology, V.T. and M.V.M.; software, A.P.; validation, V.T. and M.V.M.; formal analysis, M.V.M.; investigation, M.V.M., J.K., A.S., A.P. and J.D.; resources, A.S., J.D. and A.D.; data curation, M.V.M. and A.P.; writing—original draft preparation, M.V.M.; writing—review and editing, V.T.; visualization, M.V.M.; supervision, V.T. and A.D.; project administration, V.T. and A.D.; funding acquisition, A.D. All authors have read and agreed to the published version of the manuscript.


The EPR measurements were carried out within the SAFMAT infrastructure at FZU CAS. A. S. appreciates the Czech Science Foundation [grant number 20-21864S]. Magnetic properties were measured in the Materials Growth and Measurement Laboratory (, accessed on 3 September 2022), which is supported within the program of Czech Research Infrastructures [project number LM2018096]. CzechNanoLab project LM2018110, funded by MEYS CR, is gratefully acknowledged for the financial support of the SEM measurements. This work was supported in part by the Operational Program Research, Development, and Education financed by the European Structural and Investment Funds and the Czech Ministry of Education and Youth and Sports Project SOLID21 CZ.02.1.01/0.0/0.0/16_019/0000760.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.


V.T. kindly thanks E. Kotomin and R. Yusupov for valuable discussions. M.V.M. thanks F. Borodavka, A. Lynnik, and V. Gaertnerova for their help in conducting and processing experimental results.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.


  1. Ellialtioglu, S.; Wolfram, T.; Heurich, V.E. Surface states on n-type SrTiO3. Solid State Commmu. 1978, 27, 321–324. [Google Scholar] [CrossRef]
  2. Pemmaraju, C.D.; Sanvito, S. Ferromagnetism Driven by Intrinsic Point Defects in HfO2. Phys. Rev. Lett. 2005, 94, 217205. [Google Scholar] [CrossRef] [PubMed]
  3. Shein, I.R.; Ivanovskii, A.L. First principle prediction of vacancy-induced magnetism in non-magnetic perovskite SrTiO3. Phys. Lett. A 2007, 371, 155–159. [Google Scholar] [CrossRef]
  4. Pan, H.; Yi, J.B.; Shen, L.; Wu, R.Q.; Yang, J.H.; Lin, J.Y.; Feng, Y.P.; Ding, J.; Van, L.H.; Yin, J.H. Room-Temperature Ferromagnetism in Carbon-Doped ZnO. Phys. Rev. Lett. 2007, 99, 127201. [Google Scholar] [CrossRef] [PubMed]
  5. Ivanovskii, A.L. Magnetic effects induced by sp impurities and defects in nonmagnetic sp materials. Phys. Inorg. Chem. 2007, 177, 1083–1105. [Google Scholar]
  6. Bannikov, V.V.; Shein, I.R.; Kozhevnikov, V.L.; Ivanovskii, A.L. Magnetism without magnetic ions in non-magnetic perovskites SrTiO3, SrZrO3 and SrSnO3. J. Magn. Magn. Mater. 2008, 320, 936–942. [Google Scholar] [CrossRef]
  7. Máca, F.; Kudrnovský, J.; Drchal, V.; Bouzerar, G. Magnetism without magnetic impurities in ZrO2 oxide. Appl. Phys. Lett. 2008, 92, 212503. [Google Scholar] [CrossRef]
  8. Ye, X.J.; Zhong, W.; Xu, M.H.; Qi, X.S.; Au, C.T.; Du, Y.W. The magnetic property of carbon-doped TiO2. Phys. Lett. A 2009, 373, 3684–3687. [Google Scholar] [CrossRef]
  9. Hou, Z.; Terakura, K. Defect states induced by oxygen vacancies in cubic SrTiO3: First-Principles Calculations. J. Phys. Soc. Jpn. 2010, 79, 114704. [Google Scholar] [CrossRef]
  10. Li, N.; Yao, K.L. The electronic and optical properties of carbon-doped SrTiO3: Density functional characterization. AIP Adv. 2012, 2, 032135. [Google Scholar] [CrossRef]
  11. Zeng, H.; Wu, M.; Wang, H.-Q.; Zheng, J.-C.; Kang, J. Tuning the magnetic and electronic properties of strontium titanate by carbon doping. Front. Phys. 2021, 16, 43501. [Google Scholar] [CrossRef]
  12. Guo, Y.Y.; Liu, H.M.; Yu, D.P.; Liu, J.-M. Ferroelectricity and superparamagnetism in Sr/Ti nonstoichiometric SrTiO3. Phys. Rev. B 2012, 85, 104108. [Google Scholar] [CrossRef]
  13. Bahoosh, S.G.; Wesselinowa, J.M. The origin of magnetism in perovskite ferroelectric ABO3 nanoparticles (A = K, Li; B = Ta, Nb or A = Ba, Sr, Pb; B = Ti). J. Appl. Phys. 2012, 112, 053907. [Google Scholar] [CrossRef]
  14. Zhang, Y.; Hu, J.; Cao, E.; Sun, L.; Qin, H. Vacancy induced magnetism in SrTiO3. J. Magn. Magn. Mater. 2012, 324, 1770–1775. [Google Scholar] [CrossRef]
  15. Lin, C.; Demkov, A.A. Electron correlation in oxygen vacancy in SrTiO3. Phys. Rev. Lett. 2013, 111, 217601. [Google Scholar] [CrossRef]
  16. Zhang, S.; Guo, D.; Wang, M.; Javed, M.S.; Hu, C. Magnetism in SrTiO3 before and after UV irradiation. Appl. Surf. Sci. 2015, 335, 115–120. [Google Scholar] [CrossRef]
  17. Choi, H.; Song, J.D.; Lee, K.-R.; Kim, S. Correlated Visible-Light Absorption and Intrinsic Magnetism of SrTiO3 Due to oxygen deficiency: Bulk or surface effect. Inorg. Chem. 2015, 54, 3759–3765. [Google Scholar] [CrossRef]
  18. Liu, Y.; Zhou, W.; Wu, P. Tuning of magnetism of SrTiO3 by site-specific doping. Mater. Chem. Phys. 2015, 160, 80–86. [Google Scholar] [CrossRef]
  19. Altmeyer, M.; Jeschke, H.O.; Hijano-Cubelos, O.; Martins, C.; Lechermann, F.; Koepernik, K.; Santander-Syro, A.F.; Rozenberg, M.J.; Valentí, R.; Gabay, M. Magnetism, spin texture, and in-gap states: Atomic Specialization at the Surface of Oxygen-Deficient SrTiO3. Phys. Rev. Lett. 2016, 116, 157203. [Google Scholar] [CrossRef]
  20. Coey, J.M.D.; Venkatesan, M.; Stamenov, P. Surface magnetism of strontium titanate. J. Phys. Condens. Matter 2016, 28, 485001. [Google Scholar] [CrossRef]
  21. Brovko, O.O.; Tosatti, E. Controlling the magnetism of oxygen surface vacancies in SrTiO3 through charging. Phys. Rev. Mat. 2017, 1, 044405. [Google Scholar] [CrossRef] [Green Version]
  22. Chakraborty, B.; Nandi, P.K.; Kawazoe, Y.; Ramaniah, L.M. Room-temperature d0 ferromagnetism in carbon-doped Y2O3 for spintronic applications: A density functional theory study. Phys. Rev. B 2018, 97, 18441. [Google Scholar] [CrossRef]
  23. Choi, M.; Oba, F.; Tanaka, I. Role of Ti antisitelike defects in SrTiO3. Phys. Rev. Lett. 2009, 103, 185502. [Google Scholar] [CrossRef] [PubMed]
  24. Sundaresan, A.; Bhargavi, R.; Rangarajan, N.; Siddesh, U.; Rao, C.N.R. Ferromagnetism as a universal feature of nanoparticles of the otherwise nonmagnetic oxides. Phys. Rev. B 2006, 74, 161306. [Google Scholar] [CrossRef]
  25. Sundaresan, A.; Rao, C.N.R. Ferromagnetism as a universal feature of inorganic nanoparticles. Nano Today 2008, 4, 96–106. [Google Scholar] [CrossRef]
  26. Rice, W.D.; Ambwani, P.; Bombeck, M.; Thompson, J.D.; Haugstad, G.; Leighton, C.; Crooker, S.A. Persistent optically induced magnetism in oxygen-deficient strontium titanate. Nat. Mater. 2014, 13, 481–487. [Google Scholar] [CrossRef]
  27. Eliseev, E.A.; Morozovska, A.N.; Glinchuk, M.D.; Blinc, R. Anion vacancy-driven magnetism in incipient ferroelectric SrTiO3 and KTaO3 nanoparticles. J. Appl. Phys. 2011, 109, 094105. [Google Scholar] [CrossRef]
  28. Ishii, M.; Ohta, D.; Uehara, M.; Kimishima, Y. Vacancy-induced ferromagnetism in nano BaTiO3. Procedia Eng. 2012, 36, 578–582. [Google Scholar] [CrossRef]
  29. Golovina, I.S.; Kolesnik, S.P.; Bryksa, V.P.; Strelchuk, V.V.; Yanchuk, I.B.; Geifman, I.N.; Khainakov, S.A.; Svechnikov, S.V.; Morozovska, A.N. Defect driven ferroelectricity and magnetism in nanocrystalline KTaO3. Phys. B Condens. Matter 2012, 407, 614–623. [Google Scholar] [CrossRef]
  30. Wang, Y.F.; Shao, Y.C.; Hsieh, S.H.; Chang, Y.K.; Yeh, P.H.; Hsueh, H.C.; Chiou, J.W.; Wang, H.T.; Ray, S.C.; Tsai, H.M.; et al. Origin of magnetic properties in carbon implanted ZnO nanowires. Sci. Rep. 2018, 8, 7758. [Google Scholar] [CrossRef]
  31. Typek, J.; Guskos, N.; Zlnierkiewicz, G.; Pilarska, M.; Guskos, A.; Kusiak-Nejman, E.; Morawski, A.W. Magnetic properties of TiO2/graphitic carbon nanocomposites. Rev. Adv. Mater. Sci. 2019, 58, 107–122. [Google Scholar] [CrossRef] [Green Version]
  32. Zhang, Y.; Wang, J.; Sahoo, M.; Shimada, T.; Kitamura, T. Mechanical control of magnetism in oxygen deficient perovskite SrTiO3. Phys. Chem. Chem. Phys. 2015, 17, 27136–27144. [Google Scholar] [CrossRef] [PubMed]
  33. Mangalam, R.V.K.; Ray, N.; Waghmare, U.V.; Sundaresan, A.; Rao, C.N.R. Multiferroic properties of nanocrystalline BaTiO3. Solid State Commun. 2009, 149, 1–5. [Google Scholar] [CrossRef]
  34. Chang, C.-W.; Hu, C. Graphene oxide-derived carbon-doped SrTiO3 for highly efficient photocatalytic degradation of organic pollutants under visible light irradiation. Chem. Eng. J. 2020, 383, 123116. [Google Scholar] [CrossRef]
  35. Wang, J.; Li, H.; Li, H.; Yin, S.; Sato, T. Preparation and photocatalytic activity of visible light-active sulfur and nitrogen co-doped SrTiO3. Solid State Sci. 2009, 11, 182–188. [Google Scholar] [CrossRef]
  36. Sarkar, D.; Chu, M.; Cho, S.-J.; Kim, Y.I.; Basu, B. Synthesis and Morphological Analysis of Titanium Carbide Nanopowder. J. Am. Ceram. Soc. 2009, 92, 2877–2882. [Google Scholar] [CrossRef]
  37. Dewan, M.A.R.; Zhang, G.; Ostrovski, O. Carbothermal reduction of titania in different gas atmospheres. Metall. Mater. Trans. B 2009, 40, 62–69. [Google Scholar] [CrossRef]
  38. Chen, Y.; Deng, Y.; Zhang, H.; Wang, L.; Ma, J. A Novel and Simple Route to Synthesis Nanocrystalline Titanium Carbide via the Reaction of Titanium Dioxide and Different Carbon Source. Mater. Sci. Appl. 2011, 2, 1622–1626. [Google Scholar] [CrossRef]
  39. Strizhkov, B.V.; Lapitzkii, A.V.; Vlasov, L.G.; Tsvetkov, A.I. Produktion of titanium oxalates of divalent metals and a physical-chemical study of their thermal decomposition. Dokl. Akad. Nauk SSSR 1960, 133, 1347–1349. [Google Scholar]
  40. Hashishin, T.; Yamamoto, T.; Ohyanagi, M.; Munir, Z.A. Simultaneous synthesis and densification of titanium oxycarbide, Ti(C,O), through gas–solid combustion. J. Am. Ceram. Soc. 2003, 86, 2067–2073. [Google Scholar] [CrossRef]
  41. Miller, D.N.; Azad, A.K.; Delpouve, H.; Quazuguel, L.; Zhou, J.; Sinha, A.; Wormald, P.; Irvine, J.T.S. Studies on the crystal structure, magnetic and conductivity properties of titanium oxycarbide solid solution (TiO1−xCx). J. Mater. Chem. A 2016, 4, 5730–5736. [Google Scholar] [CrossRef]
  42. Habibi, S.; Jamshidi, M. Sol–gel synthesis of carbon-doped TiO2 nanoparticles based on microcrystalline cellulose for efficient photocatalytic degradation of methylene blue under visible light. Environ. Technol. 2020, 41, 3233–3247. [Google Scholar] [CrossRef] [PubMed]
  43. Makarova, M.; Dejneka, A.; Franc, J.; Drahokoupil, J.; Jastrabik, L.; Trepakov, V. Soft chemistry preparation methods and properties of strontium titanate nanoparticles. Opt. Mater. 2010, 32, 803–806. [Google Scholar] [CrossRef]
  44. Makarova, M.; Drahokoupil, J.; Bykov, P.; Dejneka, A.; Dlabacek, Z.; Jastrabik, L.; Trepakov, V.; Sazama, P.; Franc, J.; Okawa, Y.; et al. Size effect on the structure and optical properties in nanocrystalline SrTiO3. e-J. Surf. Sci. Nanotechnol. 2012, 10, 406–410. [Google Scholar] [CrossRef]
  45. Lee, J.S.; Khim, Z.G.; Park, Y.D.; Norton, D.P.; Theodoropoulou, N.A.; Hebard, A.F.; Budai, J.D.; Boatner, L.A.; Pearton, S.J.; Wilson, R.G. Magnetic properties of Co- and Mn-implanted BaTiO3, SrTiO3 and KTaO3. Solid-State Elect. 2003, 47, 2225–2230. [Google Scholar] [CrossRef]
  46. Berciu, M.; Bhatt, R.N. Effects of Disorder on Ferromagnetism in Diluted Magnetic Semiconductors. Phys. Rev. Lett. 2001, 87, 107203. [Google Scholar] [CrossRef] [PubMed]
  47. Goodenough, J.B.; Zhou, J.-S. Localized to itinerant electronic transition in transition-metal oxides with the perovskite structure. Chem. Mater. 1998, 10, 2980–2993. [Google Scholar] [CrossRef]
  48. Goodenough, J.B. Electronic and ionic transport properties and other physical aspects of perovskites. Rep. Prog. Phys. 2004, 67, 1915–1993. [Google Scholar] [CrossRef]
  49. Reyes-Garcia, E.A.; Sun, Y.; Reyes-Gil, K.R.; Raftery, D. Solid-state NMR and EPR analysis of carbon-doped titanium dioxide photocatalysts (TiO(2−)(x)C(x)). Solid State Nucl. Magn. Reson. 2009, 35, 74–81. [Google Scholar]
  50. Gurlo, A.; Ivanovskaya, M.; Pfau, A.; Weimar, U.; Gospel, W. Sol-gel prepared In2O3 thin films. Thin Solid Films 1997, 307, 288–293. [Google Scholar] [CrossRef]
  51. Minnekhanov, A.A.; Deygen, D.M.; Konstantinova, E.A.; Vorontsov, A.S.; Kashkarov, P.K. Paramagnetic properties of carbon-doped titanium dioxide. Nanoscale Res. Lett. 2012, 7, 333. [Google Scholar] [CrossRef] [PubMed]
  52. Konstantinova, E.A.; Kokorin, A.I.; Sakthivel, S.; Kisch, H.; Lips, K. Carbon-doped titanium dioxide: Visible light photocatalysis and EPR investigation. Chimia 2007, 61, 810–814. [Google Scholar] [CrossRef]
  53. Carrasco, J.; Illas, F.; Lopez, N.; Kotomin, E.A.; Zhukovskii, Y.F.; Piskunov, S.; Maier, J.; Hermansson, K. First principal simulations of F centers in cubic SrTiO3. Phys. Status Solidi (C) 2005, 2, 153–158. [Google Scholar] [CrossRef]
  54. Carrasco, J.; Illas, F.; Lopez, N.; Kotomin, E.A.; Zhukovskii, Y.F.; Evarestov, R.A.; Mastrikov, Y.A.; Piskunov, S.; Maier, J. First-principles calculations of the atomic and electronic structure of F centers in the bulk and on the (001) surface of SrTiO3. Phys. Rev. B 2006, 73, 064196. [Google Scholar] [CrossRef]
  55. Alexandrov, V.E.; Kotomin, E.A.; Maier, J.; Evarestov, R.A. First-principles study of bulk and surface oxygen vacancies in SrTiO3 crystal. Eur. Phys. J. B 2009, 72, 53–57. [Google Scholar] [CrossRef]
  56. Eglitis, R.I.; Kotomin, E.A.; Popov, A.I.; Kruchinin, S.P.; Jia, R. Comparative ab initio calculations of SrTiO3, BaTiO3, PbTiO3, and SrZrO3 (001) and (111) surfaces as well as oxygen vacancies. Low Temp. Phys. 2022, 48, 87–96. [Google Scholar] [CrossRef]
  57. Makarova, T.L. Magnetic properties of carbon structures. Semiconductors 2004, 38, 615–638. [Google Scholar] [CrossRef]
  58. Zitkova, J.; Zdansky, K.; Sroubek, Z. The electronic paramagnetic resonance in reduced BaTiO3 crystals. Czechoslov. J. Phys. B 1967, 17, 636–641. [Google Scholar] [CrossRef]
  59. Engelen, P.P.J.V.; Hennig, J.C.M. Electronic structure of Ti3+ centre in SrTiO3. Phys. Lett. A 1967, 25, 733–735. [Google Scholar] [CrossRef]
  60. Muller, K.A. Resonance paramagnetique du Cr3+ dans des monocristaux de SrTiO3. Arch. Sci. 1958, 11, 150–152. [Google Scholar]
  61. Mueller, K.A. Paramagnetic resonance and optical absorption of transition element ions in SrTiO3 and LaAlO2. In Proceedings of the First International Conference on Paramagnetic Resonance, Jerusalem, Israel, 16–20 July 1962; Low, W., Ed.; Academic Press: New York, NY, USA, 1963; pp. 17–43. [Google Scholar]
  62. Glinchuk, M.D.; Bykov, I.P.; Slipenyuk, A.M.; Laguta, V.V.; Jastrabik, L. ESR study of impurities in strontium titanate films. Phys. Solid State 2001, 43, 841–844. [Google Scholar] [CrossRef]
  63. Dashdorj, J.; Zvanut, M.E.; Stanley, L.J. Iron-related defect levels in SrTiO3 measured by photoelectron paramagnetic resonance spectroscopy. J. Appl. Phys. 2010, 107, 083513. [Google Scholar] [CrossRef]
  64. Trepakov, V.A.; Kapphan, S.A.; Bednorz, G.; Gregora, I.; Jastrabik, L. Dielectric-related R- zero-phonon emission line shift of Cr3+ in SrTiO3:Ca. Ferroelectrics 2004, 304, 83–86. [Google Scholar] [CrossRef]
  65. Brandão, F.D.; Pinheiro, M.V.B.; Ribeiro, G.M.; Medeiros-Ribeiro, G.; Krambrock, K. Identification of two light-induced charge states of the oxygen vacancy in single-crystalline rutile TiO2. Phys. Rev. B 2009, 80, 235204. [Google Scholar] [CrossRef]
  66. Brant, A.T.; Golden, E.M.; Giles, N.C.; Yang, S.; Sarker, M.A.R.; Watauchi, S.; Nagao, M.; Tanaka, I.; Tryk, D.A.; Manivannan, A.; et al. Triplet ground state of the neutral oxygen-vacancy donor in rutile TiO2. Phys. Rev. B 2014, 89, 115206. [Google Scholar] [CrossRef]
Figure 1. Scheme of the nanopowders synthesis. Powder colors are given for reference.
Figure 1. Scheme of the nanopowders synthesis. Powder colors are given for reference.
Crystals 12 01275 g001
Figure 2. XRD spectra of the fine and coarse nanopowders with (Cf, Cc) and without carbon doping (Rf, Rc). Reference (If) STO powder curve and standard peaks of STO (black bars) are given for comparison.
Figure 2. XRD spectra of the fine and coarse nanopowders with (Cf, Cc) and without carbon doping (Rf, Rc). Reference (If) STO powder curve and standard peaks of STO (black bars) are given for comparison.
Crystals 12 01275 g002
Figure 3. SEM image of (a) Cf nanoparticles and their as-grown aggregates (inset), (b) Rf, (c) OCf, and (d) Cc nanoparticles. (e) Particle size distributions for fine and coarse nanopowders.
Figure 3. SEM image of (a) Cf nanoparticles and their as-grown aggregates (inset), (b) Rf, (c) OCf, and (d) Cc nanoparticles. (e) Particle size distributions for fine and coarse nanopowders.
Crystals 12 01275 g003
Figure 4. (a) Magnetization loops at T = 10 K and at 300 K (Insert), (b) temperature dependence of the magnetization, and (c) saturation magnetization on specific surface area at 10 K for reduced (open squares) and oxidized (solid squares) powders. Lines are only guides for the eye.
Figure 4. (a) Magnetization loops at T = 10 K and at 300 K (Insert), (b) temperature dependence of the magnetization, and (c) saturation magnetization on specific surface area at 10 K for reduced (open squares) and oxidized (solid squares) powders. Lines are only guides for the eye.
Crystals 12 01275 g004
Figure 5. (a) Broadband and (bd) short range EPR spectra of C-doped reduced (Cf and Cc) and undoped reduced (Rf and Rc) STO nanopowders taken at room temperature (a,b) and at 25 K (c,d). The spectrum of fine oxidized carbon-doped powder (OCf) is given for comparison. For all the figures, sample mass was fixed as 20 mg, with the exception of Figure 5a.
Figure 5. (a) Broadband and (bd) short range EPR spectra of C-doped reduced (Cf and Cc) and undoped reduced (Rf and Rc) STO nanopowders taken at room temperature (a,b) and at 25 K (c,d). The spectrum of fine oxidized carbon-doped powder (OCf) is given for comparison. For all the figures, sample mass was fixed as 20 mg, with the exception of Figure 5a.
Crystals 12 01275 g005
Table 1. Sample description and notations.
Table 1. Sample description and notations.
Powder DescriptionsDesignation
SrTiO3 initial
SrTiO(C2O4)2 precursor annealed at 700 °C in air
SrTiO3 fine reduced,
If annealed at 700 °C in H2/Ar flow,
SrTiO3 coarse reduced
If annealed at 1000 °C in H2/Ar flow
SrTiO3 fine C-doped
SrTiO(C2O4)2 precursor annealed at 700 °C in H2/Ar flow
SrTiO3 white
Cf oxidized at 700 °C in air
SrTiO3 coarse C-doped
SrTiO(C2O4)2 precursor annealed at 1000 °C in H2/Ar flow
Table 2. Lattice constant a and particle size derived from XRD and SEM data.
Table 2. Lattice constant a and particle size derived from XRD and SEM data.
Designationa, ÅParticle Size, nm
Table 3. Ranking from larger to lower magnitudes of room temperature magnetization (Figure 4a,b) and EPR A-line intensity (Figure 5b) in the powders under study.
Table 3. Ranking from larger to lower magnitudes of room temperature magnetization (Figure 4a,b) and EPR A-line intensity (Figure 5b) in the powders under study.
MagnetizationA–EPR Line Intensity
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Makarova, M.V.; Prokhorov, A.; Stupakov, A.; Kopeček, J.; Drahokoupil, J.; Trepakov, V.; Dejneka, A. Synthesis and Magnetic Properties of Carbon Doped and Reduced SrTiO3 Nanoparticles. Crystals 2022, 12, 1275.

AMA Style

Makarova MV, Prokhorov A, Stupakov A, Kopeček J, Drahokoupil J, Trepakov V, Dejneka A. Synthesis and Magnetic Properties of Carbon Doped and Reduced SrTiO3 Nanoparticles. Crystals. 2022; 12(9):1275.

Chicago/Turabian Style

Makarova, Marina V., Andrey Prokhorov, Alexander Stupakov, Jaromir Kopeček, Jan Drahokoupil, Vladimir Trepakov, and Alexander Dejneka. 2022. "Synthesis and Magnetic Properties of Carbon Doped and Reduced SrTiO3 Nanoparticles" Crystals 12, no. 9: 1275.

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop